A Three-Dimensional Deep Convolutional Neural Network for Automatic Segmentation and Diameter Measurement of Type B Aortic Dissection

다국어 초록 (Multilingual Abstract)

Objective: To provide an automatic method for segmentation and diameter measurement of type B aortic dissection (TBAD).
Materials and Methods: Aortic computed tomography angiographic images from 139 patients with TBAD were consecutively collected. We implemented a deep learning method based on a three-dimensional (3D) deep convolutional neural (CNN) network, which realizes automatic segmentation and measurement of the entire aorta (EA), true lumen (TL), and false lumen (FL). The accuracy, stability, and measurement time were compared between deep learning and manual methods. The intra- and inter-observer reproducibility of the manual method was also evaluated.
Results: The mean dice coefficient scores were 0.958, 0.961, and 0.932 for EA, TL, and FL, respectively. There was a linear relationship between the reference standard and measurement by the manual and deep learning method (r = 0.964 and 0.991, respectively). The average measurement error of the deep learning method was less than that of the manual method (EA, 1.64% vs. 4.13%; TL, 2.46% vs. 11.67%; FL, 2.50% vs. 8.02%). Bland-Altman plots revealed that the deviations of the diameters between the deep learning method and the reference standard were -0.042 mm (-3.412 to 3.330 mm), -0.376 mm (-3.328 to 2.577 mm), and 0.026 mm (-3.040 to 3.092 mm) for EA, TL, and FL, respectively. For the manual method, the corresponding deviations were -0.166 mm (-1.419 to 1.086 mm), -0.050 mm (-0.970 to 1.070 mm), and -0.085 mm (-1.010 to 0.084 mm). Intra- and inter-observer differences were found in measurements with the manual method, but not with the deep learning method. The measurement time with the deep learning method was markedly shorter than with the manual method (21.7 ± 1.1 vs. 82.5 ± 16.1 minutes, p < 0.001).
Conclusion: The performance of efficient segmentation and diameter measurement of TBADs based on the 3D deep CNN was both accurate and stable. This method is promising for evaluating aortic morphology automatically and alleviating the workload of radiologists in the near future.

참고문헌 (Reference)

1 Hagan PG, "The International Registry of Acute Aortic Dissection (IRAD): new insights into an old disease" 283 : 897-903, 2000

2 Ben Ayed I, "TRIC: trust region for invariant compactness and its application to abdominal aorta segmentation" 17 (17): 381-388, 2014

3 Behrens T, "Robust segmentation of tubular structures in 3-D medical images by parametric object detection and tracking" 33 : 554-561, 2003

4 Rengier F, "Reliability of semiautomatic centerline analysis versus manual aortic measurement techniques for TEVAR among non-experts" 42 : 324-331, 2011

5 Alom MZ, "Recurrent residual U-Net for medical image segmentation" 6 : 014006-, 2019

6 Wang G, "Machine learning will transform radiology significantly within the next 5 years" 44 : 2041-2044, 2017

7 Tajik AJ, "Machine learning for echocardiographic imaging:embarking on another incredible journey" 68 : 2296-2298, 2016

8 Kaji S, "Long-term prognosis of patients with type B aortic intramural hematoma" 108 (108): II307-II311, 2003

9 Drozdzal M, "Learning normalized inputs for iterative estimation in medical image segmentation" 44 : 1-13, 2018

10 Sternbergh WC 3rd, "Influence of endograft oversizing on device migration, endoleak, aneurysm shrinkage, and aortic neck dilation: results from the Zenith Multicenter Trial" 39 : 20-26, 2004

11 Novelline RA, "Helical CT in emergency radiology" 213 : 321-339, 1999

12 Li X, "H-DenseUNet:hybrid densely connected UNet for liver and tumor segmentation From CT volumes" 37 : 2663-2674, 2018

13 Cao L, "Fully automatic segmentation of type B aortic dissection from CTA images enabled by deep learning" 121 : 108713-, 2019

14 Rengier F, "Development of in vivo quantitative geometric mapping of the aortic arch for advanced endovascular aortic repair: feasibility and preliminary results" 22 : 980-986, 2011

15 Chartrand G, "Deep learning: a primer for radiologists" 37 : 2113-2131, 2017

16 Sahiner B, "Deep learning in medical imaging and radiation therapy" 46 : e1-e36, 2019

17 이준구, "Deep Learning in Medical Imaging: General Overview" 대한영상의학회 18 (18): 570-584, 2017

18 Pereira S, "Brain tumor segmentation using convolutional neural networks in MRI images" 35 : 1240-1251, 2016

19 Moeskops P, "Automatic segmentation of MR brain images with a convolutional neural network" 35 : 1252-1261, 2016

20 Herment A, "Automated segmentation of the aorta from phase contrast MR images: validation against expert tracing in healthy volunteers and in patients with a dilated aorta" 31 : 881-888, 2010

21 Sedghi Gamechi Z, "Automated 3D segmentation and diameter measurement of the thoracic aorta on non-contrast enhanced CT" 29 : 4613-4623, 2019

22 Sommer T, "Aortic dissection: a comparative study of diagnosis with spiral CT, multiplanar transesophageal echocardiography, and MR imaging" 199 : 347-352, 1996

23 Müller-Eschner M, "Accuracy and variability of semiautomatic centerline analysis versus manual aortic measurement techniques for TEVAR" 45 : 241-247, 2013

24 Tian Y, "A vessel active contour model for vascular segmentation" 2014 : 106490-, 2014

25 Wang G, "A perspective on deep imaging" 4 : 8914-8924, 2016

26 Erbel R, "2014 ESC guidelines on the diagnosis and treatment of aortic diseases: document covering acute and chronic aortic diseases of the thoracic and abdominal aorta of the adult. The Task Force for the Diagnosis and Treatment of Aortic Diseases of the European Society of Cardiology (ESC)" 35 : 2873-2926, 2014